C E R T

CENTRE FOR ECONOMIC REFORM AND TRANSFORMATION

Department of Economics, Heriot-Watt University, Riccarton, Edinburgh, EH14 4AS
Tel: 0131 451 3623 Fax: 0131 451 3498 E-Mail: [email protected]
World-Wide Web: http://www.hw.ac.uk/ecoWWW/cert/certhp.htm
Measuring Progress in Transition
and towards EU Accession
A comparison of manufacturing firms
in Poland, Romania and Spain
Wendy Carlin
Department of Economics, University College London
[email protected]
Saul Estrin
Department of Economics, London Business School
[email protected]
Mark Schaffer
Centre for Economic Reform and Transformation
Heriot-Watt University
[email protected]
Discussion Paper No. 99/02. J anuary 1999
Abstract
This paper provides new evidence on progress in transition and the readiness of
enterprises for accession to the EU using a detailed survey administered to
approximately 200 manufacturing firms in each of Poland, Romania and Spain. A
major innovation is the use of a market economy and member country of the EU
Spain as a benchmark against which to measure progress in transition. The paper
finds that new private firms (firms established as private ab initio) in both Poland
and Romania are growing the fastest, but on measures of integration and investment,
it is Polish ab initio private firms and privatized firms that look most similar to the
Spanish. Polish state-owned firms, and Romanian state-owned, privatized and (to a
lesser extent) ab initio private firms more often lag behind. With respect to
compliance with EU directives, Poland tends to lag behind Spain but lies significantly
ahead of Romania. Levels of awareness of and compliance with EU directives did not
vary with ownership type amongst the Eastern European firms. Progress in transition
at the country level seems to be consistent with improvements in compliance with the
major components of the acquis.
Acknowledgments
Paper prepared for the Chief Economist s Office, European Bank for Reconstruction and Development. Assistance
from Savvas Kyriakides and Junior Davis in the preparation of the survey questionnaire, and the comments and
suggestions of the participants at a workshop at the European Commission on 9 December 1998, are gratefully
acknowledged. The usual disclaimer applies.
2
1. Introduction
This paper provides new evidence on the readinessof enterprises from two contrasting
pre-accession countries for accession to the EU. A detailed survey was administered to
approximately 200 firms in the manufacturing sector in each of Poland and Romania in
the summer of 1998. In order to provide a benchmark from a large EU member
country with a relatively low per capita GDP, 200 firms in Spain were also surveyed.
The requirements for accession are the following: (a) the existence of a market
economy; (b) the capacity to take over and implement the acquis; and (c) the capacity
to withstand competitive pressures. An enterprise level survey can provide information
on some aspects of these accession requirements. The objective of the survey was to
answer three questions: (i) How far have both privatized and state-owned firms moved
in the direction of the general characteristics of firms in a market economy? How
closely do new private firms those established ab initio resemble those in a market
economy? (ii) How well do enterprises currently meet a set of key administrative
requirements for accession (aspects of the acquis)? We sought to discover the extent
of existing compliance; and the estimated costs associated with attaining current
compliance levels and plans for improving compliance. (iii) How do the performance
and characteristics of enterprises compare with those of firms in an EU member state?
There is no direct measure of the ability of enterprises to withstand competitive
pressures. But we can look at the relationship between performance and the observable
characteristics of corporate governance and competitive environment in the accession
countries and compare them with the benchmark country.
At the outset of transition, the Eastern European countries were characterized by a
pattern of trade dominated by sales within the former-CMEA area; the dominance of
state ownership and absence of both private and foreign ownership; the absence of
product market competition as a determinant of resource allocation, x-efficiency and
enterprise survival; and the virtual absence of small and medium-sized firms. The
enterprises in the region produced products of poor quality using equipment that was
frequently technologically obsolete and that did not meet modern environmental
standards. Overmanning was also widespread. Major changes along all these
dimensions toward convergence with the characteristics of market economies have
been recorded over the course of transition. Many enterprise level surveys have been
conducted over the past seven or eight years with the objective of describing the
pattern of change and identifying the proximate causes of changes in performance. A
problem with these surveys has been the absence of a direct benchmark by which
convergence towards a market economy can be assessed. Therefore we have included
an EU comparator country in this study both as a contribution to the transition debate
and in order to be able to provide new information for the accession debate.
The transition literature suggests that market-oriented adjustment was characteristic of
firms of all ownership types in the first phase of reactiveor cost-orientedrestructuring
3
in the presence of the opening of competition in product markets and the withdrawal of
ex post financial support to loss-making firms. Theoretical considerations suggested
that deep or strategic restructuring would require profit-orientation most likely
delivered by private ownership with effective corporate governance. Some empirical
evidence has emerged to support this contention.
1
The rapidity with which firms in Eastern Europe reoriented their sales from former
CMEA to European Union markets has been one of the most remarkable features of
the transition. One indicator of how well integrated East European firms are in the
European market is the degree of similarity between the exporting activity of Eastern
European firms and their EU counterparts. Are they similarly export oriented, do they
perceive similar levels of import competition, is the quality profile of export as
compared with domestic sales similar to that of EU firms? This entails a comparison
between Spanish firms and those in Poland and Romania. To the extent that differences
persist, it is interesting to know whether they are related to standard transition
indicators. The transition literature suggests that ab initio firms will perform much like
firms in market economies, while state-owned enterprises will continue to perform
poorly across a wide range of indicators. An important indicator of progress in
transition is the extent to which former state-owned firms, now privatized, also display
the characteristics of market economy private firms.
It is also possible that country differences will overwhelm such effects. Poland entered
the transition at a higher level of development than did Romania, had a pre-existing ab
initio private sector in manufacturing, began the transition earlier, emerged from the
transition-recessionafter the elapse of fewer transition years and has experienced many
more years of growth than has been the case for Romania. It is expected that such
country differences would be reflected in performance and integration.
A major objective of the European Single Market programme (1992 Programme) was
to create a borderless market within the EU, in the sense that the conditions of export
sales from one EU member to another would be indistinguishable from sales within the
home country. By including Spain in the survey, it is possible to establish an EU-
benchmark for the gapbetween firms exporting to the EU and those confined to the
home market. Are Spanish firms that only sell in the domestic market very different
from those that export to the EU? As the single market becomes a reality, one would
expect such differences to diminish. Given that the Polish and Romanian firms do not
belong to this single market, we could form the hypothesis that there will be more
substantial differences between those firms oriented to selling into the EU market and
those selling at home within each of the East European countries. The conditions of
competition, for example, might well be different. If it were the case that the
characteristics of firms in the transition economies that were selling successfully in the

1
See, e.g., EBRD, Transition Report 1997, for an extended survey and discussion.
4
EU market were indistinguishable from those of firms selling only at home, one would
be less concerned about the capability of the latter to survive under single market
conditions.
Readiness for accession includes compliance with specific EU directives and
regulations. The survey instrument was designed to uncover the extent of compliance
of firms in Poland and Romania and also in Spain. The directives addressed were those
relating to working time, equal pay, health and safety at work, emissions monitoring
and product certification. We were interested in uncovering whether compliance levels
in these different spheres in the East European countries were a legacy from the
communist period or were related to transition indicators such as privatization and
integration with the EU through exporting or foreign ownership. An effort was made
to ascertain the cost of adjustment if any that had been needed to achieve compliance,
the kinds of costs incurred (e.g., fixed investment, training) and the extent to which
assistance had been provided by government or foreign partners in attaining
compliance.
In the first section of the paper, a brief summary of the survey design and sample
selection criteria is provided. Section 2 turns to the results of the survey and in
particular to the question of the structural and performance characteristics of
enterprises in the three countries. The objectives of this section are to find out the
extent to which there are differences between the enterprises in the transition countries
and Spain, and between enterprises in the two transition countries, and what those
differences are. For example, do Polish and Romanian firms look more like each other
or more like Spanish firms in terms of their output and employment growth, or in terms
of ownership structure state-owned, privatized, or ab initio private? Are firms in the
transition economies similar in terms of their exporting activities or are there country
or ownership differences? We organize the analysis of similarities and differences by
looking in turn at performance, international integration, governance structure and
competitive environment. Section 3 addresses the compliance issues. It begins by
describing the levels of compliance achieved in each of the three countries, explores the
possibility that there might be directives for which compliance is legacy-basedrather
than transition-based and investigates the link between the different transition
indicators and compliance. In Section 4, attention is focused on what can be learned
from this survey about the determinants of readinessfor accession and the perceptions
of firms about how accession might affect them. We present a comparison between the
benefits and costs expected by the firms in pre-accession countries from membership of
the EU with the retrospective evaluation of benefits and costs by the Spanish firms.
Section 5 concludes with a summary of the major findings from the survey,
highlighting both expected and puzzling results.
5
2. Survey design
The survey covered 645 manufacturing firms in total: 215 from Spain, 223 from
Poland and 207 from Romania. Firms in each country were selected randomly
2
but
within certain restrictions:
Manufacturing firms were selected from the following sectors only: mechanical
engineering and transport, wood products and furniture, food processing,
clothing and footwear, chemical products, and electrical machinery. Furthermore,
firms from the following sectors were explicitly excluded: building and
construction materials; manufacture of basic metals; shipbuilding; newspapers
and other printing. There were no sectoral quotas; firms were selected randomly
from the pool of allowable manufacturing sectors.
Firms were selected from two locations in each country: Madrid and Barcelona
in Spain, Warsaw and Katowice in Poland, Bucharest and Brasov in Romania.
In each country, the size distribution of firms was to be divided into three
comparably-sized groups: small firms, employing 50150 people; medium-sized
firms, employing 150 to 500 people; and large firms, employing 500 to 5,000
people. Firms employing fewer than 50 persons or more than 5,000 were
excluded from the survey.
In Poland and Romania only, the ownership distribution of the firms included in
the survey was to be divided into three groups: majority or 100% state-owned
firms; privatized (previously state-owned) firms; and ab initio private firms. The
small firm category in each of these countries was to be divided between ab
initio firms (approximately 50%) and privatized or state-owned firms
(approximately 50%). The medium-sized firm category was to be divided as with
small firms. The large firm category was to be divided roughly equally between
privatized firms and state-owned firms.
This last feature of the sampling strategy was motivated in large part by the limited
numbers of firms to be surveyed in each country. To take two examples, without such
stratification, the Polish small firm group would have been likely to be composed
mostly of ab initio private firms, reflecting the strong growth of this sector in Poland in
the transition period; and the large firmgroup in Romania would have been likely to be
composed mostly of state-owned firms, reflecting the slow pace of privatization there.
Ab initio firms were excluded a priori from the largecategory because of the small
numbers of such firms in both the Polish and Romanian economy. We note here that
export activity was not a criteria for selection, and hence the observed distribution of
export activity is driven by the characteristics of the population of firms and not by the
sampling procedure.

2
From the databases maintained by the government statistical offices in the cases of Poland and
Romania, and from a large commercial database in the case of Spain.
6
The survey was conducted by the company MEMRB in June-September 1998. The
main survey was preceded by a pilot survey of 20-odd firms in each country in March-
April, after which revisions to the survey questionnaire were made in light of the pilot
results. Information was collected by interview, typically with one or more members of
the top management of the firm.
Table A1 in the Appendix presents the basic descriptive data for the firms surveyed.
The table also presents average and median employment by ownership category state,
privatized, ab initio for each country. As one would expect, both a priori and from
the sampling design, there is a clear correlation between ownership and size: state-
owned firms tend to be the largest, followed by privatized firms, with new private firms
the smallest. For this reason, when we test statistically the observed differences across
ownership categories, we attempt to control for size (see the Appendix for further
details). This is to try to avoid, for example, confusing the characteristics of ab initio
firms that are the result of their smaller size with those that are genuinely related to
their ownership status.
3. Structure and performance of firms: are transition firms different?
3.1 Performance
The debate about how to measure firm performance under conditions of transition is
unresolved. Rather than focusing on a single measure, we present several. The first
problem we face in our cross-country comparisons of performance is that Romania
experienced a significant setback in 1997, with industrial output declining by 6% after
two years of strong growth, and with inflation at about 150% for the year. We
therefore compare the 1996 performance of the Romanian firms in the survey with the
1997 performance for the Polish and Spanish firms.
A measure of profitability is sometimes used as a synthetic index of performance but
suffers from well known reporting and measurement problems. When inflation is high,
the interpretation of the profit to sales ratio is especially difficult because of the
upward bias introduced by historical cost accounting;
3
this is a particular problem for
Romania, whose PPI inflation was 50% in 1996, compared to1997 PPI inflation rates
of 9% for Poland and 1% for Spain. We have attempted to correct the data for the
effects of inflation,
4
but the comparisons between Romanian and other firms should

3
Profit is calculated using costs in historical or purchase prices. When inflation is high, costs are
understated relative to sales revenue because of the general increase in the price level between the date
inputs are purchased and the date the output embodying these inputs is sold.
4
Profit is re-estimated using material costs that are inflated by an average of two months inflation to
compensate for changes in the general price level during the production process. The official PPI is
used for Spain and Poland; for Romania, the implicit industrial price deflator (calculated from
nominal industrial output and the official real output index) is used.
7
still be treated with caution. Profitability (gross operating surplus
5
as a proportion of
sales revenue) is higher for the Polish ab initio firms than for Spanish firms but
otherwise Polish firms are similar to Spanish ones (see Table 1). Romanian private
firms, both privatized and ab initio private, also appear more profitable than their
Spanish counterparts, but as just noted this is not a firm finding. The interpretation of
the profit to sales ratio is also muddied by the fact that it will reflect the conditions of
product market competition.
Considerable theoretical and empirical evidence has accumulated to suggest that
adjustment to a market economy takes place in stages. Initially managers react to the
introduction of market forces and the hardening of their budget constraints by cutting
costs. One indicator of this is labour shedding. We first look at the extent to which
firms in the sample identify the existence of labour hoarding and then at the changes in
employment that are reported. Another component of early adjustment is to find
markets for the output of the firm. Whilst macroeconomic conditions exert a strong
influence over the growth of output in the country, it is instructive to look at the
pattern of growth of sales where firms are distinguished by ownership type. A second
stage of restructuring has been characterized in terms of the emergence of a strategic
orientation of firms. Observable indicators of deep restructuring are usually taken to be
investment in fixed capital or human capital or in R&D.
In the early stage of transition, excess labour in the state sector, exacerbated by the
economy-wide recession experienced by all transition economies, is typically reflected
in labour shedding by state and former state firms as managers act to contain costs in
the face of increasing competition and declining state support. As transition proceeds,
one would expect labour shedding to disappear from the privatized firms with an
eventual transition to privatized firms sharing in growth with the ab initio sector.
Somewhat surprisingly, in the survey data for 1997, the latter pattern does not yet
appear to have emerged in Poland (see Table 2). Growth of employment in ab initio
firms is rapid and indistinguishable from that of Spanish firms, whilst privatized firms
as well as SOEs continue to contract. The pattern is the same in Romania.
The explanation for why we do not yet find most privatized Polish firms expanding
their employment may, however, be specific to Poland and relate to the existence of
excess labour. A peculiarity of the Polish transition still appears to be present here. In
spite of progress on the dimensions discussed above, Polish state and privatized firms
stand out from their Romanian counterparts in their propensity to hoard labour (see
Table 3). In Romania there is still some labour hoarding in the state sector, as
measured by comparison with Spanish firms, but not elsewhere.

5
Earnings before interest, tax and depreciation.
8
There is some support from the survey for the predicted pattern in sales growth. The
set-back to transition in Romania in 1997 highlights the role of macroeconomic factors
and should be kept in mind when interpreting Table 4. Against the Spanish benchmark,
a ranking of Polish firms emerges: the growth of Polish ab initio firms is as rapid as
that of Spanish firms, with Polish privatized firms growing more slowly and SOEs
output actually shrinking (see Table 4). The relative performance of the different types
of Romanian firms is similar to that in Poland, with new private firms expanding sales
faster than privatized ones and with the sales of SOEs virtually stagnant.
The recovery of investment activity is frequently referred to in the transition literature
as an indicator of deep restructuring and of progress in transition. It is thought to
require managers who are oriented toward the future profitability and strategic
development of the firm, as compared with the initial transition task of surviving the
marketization shock. It would be expected that privatization in the presence of
effective corporate governance would be associated with higher investment. From the
survey it appears that both privatized and ab initio private firms in Poland have
substantially higher investment shares than is the case in Spain, in line with the
expected pattern as transition proceeds (see Table 5). Although ab initio Romanian
firms are investing more than Spanish ones, the situation in Romanian privatized firms
looks poor with investment well below that in the Polish privatized firms and no
different from the levels in Romanian state firms.
In both training and R&D activity, Polish and Spanish firms are alike with levels
significantly higher than Romanian firms. There is also some evidence that Polish
privatized and ab initio private firms have a significantly higher share of marketing
staff in total employment (at Spanish levels) than do state firms in Poland or all firms in
Romania.
The difference between the Polish and Romanian privatized firms appears to be that the
former are engaging in deep restructuring (as evidenced by, for example, their
investment rates) even though their workers have often been able to exert enough
influence to protect jobs, whereas the latter are still in the earlier phase of reactivelow-
investment restructuring.
3.2 I nternational integration
There are two very striking differences between the pattern of exporting behaviour of
Spanish, Polish
6
and Romanian firms. These are best understood by looking at
histograms showing the distribution of exports as a percentage of sales across firms
(see Figure 1). A glance at the histogram shows that the Spanish distribution is quite

6
In Poland, 24 of the 41 firms that export more than one-quarter of their sales to the EU are sub-
contractors for 100% of their output. (In Romania, the corresponding number of firms is five out of
46.) We tested fairly extensively for an effect of sub-contracting on performance and compliance
outcomes but were unable to uncover any significant effects.
9
different from that of the Eastern European countries. The great majority of Spanish
firms are involved in some sales abroad. By contrast, in both the Eastern European
countries there is a very marked spike at the left-hand side with between 30 and 60%
of firms (depending on country and ownership type) exporting less than 10% of sales.
Many firms do not export at all. The other notable result is that in Romania there is a
second concentration of privatized and ab initio firms exporting at least 80% of their
output. The widespread participation of Spanish firms in exports highlights the greater
extent of its integration in international trade and in the European market. Polish
privatized firms come closest to the Spanish pattern. By contrast, Romanian private
firms (and especially new private firms) display a separation between non-exporters, an
intermediate group and specialist exporters.
As would have been expected, Spanish firms export to the EU to a greater extent than
is the case for the Eastern European firms, irrespective of ownership type. In neither
Poland nor Romania does it appear that the exporting activity of state firms is
dominated by a legacy of trade links with former CMEA countries. Indeed, in Poland,
it is ab initio firms that have the strongest trade links with former CMEA countries.
This probably reflects the seizure of market opportunities by Polish entrepreneurs for
example, in successfully identifying profitable market niches in the Russian economy in
the wake of liberalization. The parallel in Romania to this group of Polish start-ups is a
dozen ab initio firms that are specialist exporters selling at least three-quarters of their
output to the EU all are clothing firms, smaller than average and doing better than
average.
If Spanish firms are operating in a single European market, one would expect that
there would be less difference between the price-quality characteristics of goods
exported and those sold at home than would be the case for firms in a market
segmented from the EU. Firms in Eastern Europe, for example, may have a domestic
market that is protected from pan-European competition and in which different
conditions of sale (e.g., product standards) apply, whereas exports must sell in head-
to-headcompetition under the same conditions with the products of EU countries.
Respondents to the survey were asked to identify the main product that was exported.
They were then asked if the same product was sold at home. For firms that sold the
same product at home and abroad, we then asked whether the exported product was of
lower, higher or the same quality. Virtually all Spanish firms did sell the same product
at home and abroad and it was of the same quality (see Table 6). Many fewer Polish
firms sold the same product at home and abroad and even fewer Romanian ones did so.
In an interesting reflection of the closer convergence of Poland to Spain, Polish private
firms, both privatized and ab initio private, tended to sell the same quality of good just
like the Spanish firms. By contrast, Polish state-owned firms looked like Romanian
firms (of all types) with a greater tendency to sell higher quality goods for export than
for the home market.
10
3.3 Governance structure
One solution to the agency problem that arises when ownership and control in a firm
are separated is concentrated ownership. The presence of a shareholder with a
substantial stake helps to overcome the standard free-rider problem in control. For this
to be a solution requires, in turn, that the large shareholder is motivated by the
objective of value maximization and that the private benefits that are derived from
ownership do not conflict with this. In short, for concentrated ownership to be
effective, the owner needs not just to have the capability of monitoring management
but also the incentive to do so in the pursuit of efficiency rather than private gain (for
example through asset expropriation). The understanding of the precise determinants
of effectivecorporate governance is far from complete even in a market economy. Our
aim here is to highlight the similarities and differences between ownership
concentration in the Spanish economy and the private sector of the transition countries.
We find that in terms of concentration, Spain looks different from the transition
countries (Table 7). Ownership is significantly more likely to be concentrated in Spain
and there is no significant difference between Poland and Romania. It is the privatized
Eastern European firms that look very different from the Spanish ones. There is less
concentrated ownership in Poland and Romania in privatized firms suggesting the
possibility of more serious monitoring problems.
It has frequently been suggested in the debates about transition that foreign ownership
represents an attractive method of privatization bringing an owner with profit
orientation and with access to finance for investment, to management expertise and to
markets in the West. It is interesting to note that foreign ownership is much more
prevalent amongst the Spanish firms than amongst either privatized or ab initio private
firms in Poland or Romania (see Table 8). The high foreign ownership of Spanish firms
reflects changes in Spain as a consequence of EU accession and of the single market
measures providing a pointer to what might be expected for the Eastern European
countries after accession. The tendency for Polish firms to be foreign-owned to a
greater extent than Romanian ones is not quite statistically significant overall although
Polish ab initio firms are more likely to be foreign-owned than Romanian privatized
ones. The role of foreign ownership in Polish start-ups is interesting. One of the
debates about the consequences of the design of privatization programmes in transition
is whether privatization to insiders could lead to the entrenchment of incumbent
managers who become owners. The question of whether insider-privatization has
inhibited the transfer of ownership both in terms of concentration and foreign
ownership for Polish privatized firms (which look different from Spanish ones on both
counts) is one that deserves closer investigation.
11
3.4 Competitive environment
The inclusion of Spain in the survey allowed for the identification of a benchmark
against which the pressure of competition felt by Eastern European firms could be
measured. Firms were asked whether, in the market for their main product, they faced
zero competitors, between one and five competitors or more than five. There was no
significant difference in the judgements of Spanish and Polish firms as to their
competitive environment (see Table 9). It is interesting if slightly puzzling that Polish
SOEs are very similar to Romanian privatized and ab initio private firms in being more
likely to face more competitors. On the other hand, Romanian SOEs are less likely to
face more competitors. One possible interpretation is that Romanian state firms are at a
very early stage of transition with limited perception of competitive pressure. By
contrast, Romanian privatized and especially ab initio private firms (leading the
Romanian transition) and Polish state firms (lagging the Polish transition) are very
aware of competition. Finally, the transition firms best adapted to the market economy
are in more settled markets and face a pattern of competition similar to that of market-
economy firms.
The findings regarding the importance attached to competition from imports are also
interesting in this regard. There is a clear split between the Eastern European firms and
the Spanish ones. Import competition is more frequently viewed as important by the
transition than the market economy firms (see Table 10). Perhaps the Spanish firms
have become less sensitive to the origin of competing products as the EU market has
become more borderless. There are no discernible differences between Polish and
Romanian firms on this measure.
3.5 Summary
Looking across the four dimensions of structure and performance, we find that there
are a number of respects in which there is a significant gap between Spanish firms on
the one hand and Eastern European ones both Polish and Romanian on the other.
This is by no means the case across the board. The differences emerge in some aspects
of governance, international engagement and competition. Spanish firms are more
likely to have a single majority owner and they are more likely to be foreign owned.
This suggests the presence of more effective corporate governance of Spanish than of
transition firms. The perception of import competition is substantially higher in the
Eastern European economies than in Spain. As noted above, this apparently counter-
intuitive result may simply reflect the progress that has been made in the single
European market, with Spanish firms ceasing to think in terms of importand domestic
competition. This interpretation is supported by the finding that hardly any Spanish
firms are not engaged in sales to other EU countries. The share of exports to the EU
by Spanish firms is significantly higher than for transition firms confirming that Spanish
firms are more integrated in the EU than those from Poland and Romania.
12
The results in this section suggest that in a number of respects Polish ab initio private
firms, and to a lesser extent Polish privatized firms and Romanian ab initio firms, are
rather similar to Spanish firms and different from Romanian state-owned and privatized
ones. One would expect to observe this pattern as transition proceeds, with private
sector firms whether their origins were in the state-owned sector or as start-ups
coming to look more like firms in long-established market economies. Clear
convergence of Poland toward Spain is revealed by the distribution of exporting
activity across firms. As vividly depicted in the histograms (Figure 1), Romania is
characterized by a set of specialist exporting firms. Such firms are absent from the
Spanish sample and relatively rare in the Polish sample.
Two other results emerge that capture differences within Eastern Europe. First, Polish
private firms have higher investment ratios than do Spanish ones. This reflects the
exploitation of Polands scope for catch-up to countries with higher productivity levels.
By contrast, investment is especially low in privatized Romanian firms indicating that
the conditions for effective catch-up have yet to be established for former SOEs.
Secondly, the well-documented phenomenon of excess labour continues to
characterize Polish firms. Whereas in Romania, labour hoarding is characteristic only
of SOEs, in Poland privatized firms that on many other counts are well on the way to
convergence with Spanish ones are still bearing this legacy from the 1980s. The excess
labour represents unutilized potential for improved competitiveness of these Polish
firms their gross profit margins (operating surplus as a proportion of sales) are similar
to Spanish firms.
4. Compliance with EU directives and regulations
The survey instrument was designed both to identify objective measures of compliance
and to gauge the perception of respondents as to their degree of compliance. There
was a close correspondence between the results from the objective measures (e.g.,
what proportion of your workforce works for more than x hours per week?) and the
subjective assessment of whether they would have satisfied an EU inspector making an
unannounced visit to check compliance.
7
The questions about compliance were set up
so that the respondent was led through the various components of the directive before
being asked to evaluate the firms level of compliance with the directive as a whole (see
the Appendix where an edited extract from the section of the questionnaire dealing
with the compliance questions is presented). From a policy perspective it is useful to
know whether levels of compliance were inherited from the pre-transition period and
whether there is any relationship between ownership type or other transition indicators
in the extent of compliance in Eastern Europe.

7
As the assessment of compliance is that of the interviewee, we have more confidence in the rankings
of compliance across directives than in absolute levels of compliance for any particular directive.
13
4.1 Equal pay directive
Compliance levels with the equal pay directive were uniformly high across all three
countries (see Table 11). There were no size or ownership effects in the compliance
achieved by the time of the survey. Firms in the Eastern European countries reported
that compliance levels had been as required by the Directive for at least five years or
since start-up. Somewhat surprisingly, current compliance levels had been achieved
relatively recently amongst the state firms with just over one-half of SOEs in Poland
and only one-quarter of those in Romania meeting directive levels in 1989 or earlier.
4.2 Working time directive
Although compliance with the working time directive is high in the Eastern European
countries with over three-quarters of firms reporting high compliance, Spain is still
ahead of Poland and Poland ahead of Romania (see Table 12). It appears that Polish ab
initio firms are still behind other Polish firms in compliance. This suggests that any
government effort in raising compliance levels would need to be directed toward start-
up firms in Poland. In Romania, compliance levels are uniform across ownership types
at a lower level than for Poland.
4.3 Occupational safety directive
In meeting the occupational safety directive, Polish firms report the same level of
compliance as Spanish ones.
8
This applies across the board irrespective of ownership
type (see Table 13). Only one-third of the Polish state firms had current levels of
compliance in 1989, which indicates that adjustment has taken place during transition.
Most privatized and ab initio firms report that current levels were achieved from
privatization or start-up or earlier. Romanian firms are well behind the others in
compliance. Interestingly, it is new start-ups in Romania that show higher compliance
levels at least as compared with SOEs. Start-up firms in both Poland and Romania
indicated that substantial changes were required in order to meet the compliance levels
specified in the directive. In neither case was outside help a significant factor in making
the necessary changes.
4.4 Product certification requirements for exports to the EU; I SO
certification
For firms outside the EU, any exports to the EU must comply with product
certification procedures. Polish firms were more aware of these requirements than
were Romanian firms (see Table 14). The second question that we wanted to pursue
was whether there was any difference in their awareness of these requirements
according to whether a firm was engaged in selling goods in the EU market. In section

8
The caveat above about absolute levels of compliance is particularly important here. The
interviewees judgement of whether the firm has satisfied the occupational safety directive could be
influenced by the average quality of safety equipment in that country; e.g., what is seen as a
satisfactory standard of equipment in Romania may not be acceptable in Spain.
14
2 above, it emerged that Polish firms appeared to be more integrated into the EU
market whereas in Romania a group of specialist exporting firms existed. Another
indicator of integration would be an awareness of certification requirements for exports
even amongst firms that did not currently export. The survey results suggest that there
is a distinction between Polish and Romanian firms along these lines. Polish firms seem
to be equally aware of the certification requirements for exports to the EU even when
they are not selling into that market. By contrast, whilst Romanian exporters were just
as conscious of the requirements as Polish firms, the level of awareness amongst non-
exporters was much lower.
Familiarity with export certification requirements does not vary according to the type
of ownership of the firm or whether or not ownership is concentrated it seems to be a
country effect. Firms that claimed to be familiar with the export certification
requirements were also asked how much of their output meets these requirements. The
share of output meeting this standard is highest in Polish privatized and ab initio
private firms and Romanian privatized ones. Romanian start-ups and state firms were
significantly lower. This highlights the greater depth of convergence to EU standards
across the Polish private sector and underlines the specialization of Romanian
privatized firms in exporting activity. To pursue this question further, we were
interested in finding out the cross-country differences in ISO quality standards
certification as another proxy for EU product standards in the internal market.
Spanish firms were significantly more likely to have ISO9000 certification with nearly
two-thirds of firms (controlling for size of firm) complying (Table 15). By contrast less
than one-quarter of firms in Poland and Romania had ISO certification. There was no
difference between Polish and Romanian firms, no ownership differences and no
tendency for exporting firms to have certification more often than non-exporting firms.
However, one intriguing pattern does emerge. In both Poland and Romania, there is a
tendency for firms that dominate their market (i.e., that identify no competitors for
their main product) to be more likely to have ISO-certification. In Romania, for
example, of the 28 firms that say they dominate the market, over half have ISO
certification. An appealing hypothesis is that these firms are sub-contractors with the
foreign purchasing firm assisting with ISO certification. This hypothesis is refuted by
the data over half the firms are state-owned and only one-quarter do any sub-
contracting at all.
Polish firms that achieved current compliance levels for export certification in 1995 or
later indicated that the changes required in order to do so involved fixed investment,
training and changes in the production process in equal measure. The corresponding
Romanian firms less frequently mentioned training. Most Polish and Romanian firms
received no outside help in achieving compliance: a handful of firms in each country
mentioned that they had had some foreign help. There was a somewhat higher level of
outside help reported for the achievement of ISO certification with about one-half of
Polish firms reporting outside assistance, most often from a foreign customer, partner
15
or owner. There was sparse mention of government help. The Polish firms with ISO
certification virtually all mentioned the need for increased training with about one-half
noting the need for new investment and changes in the production process. Romanian
firms with ISO certification appear to have done so without outside assistance as
would be expected from their characteristics noted above.
4.5 Emissions monitoring
In a wide-ranging survey of firms from a variety of manufacturing sectors, there was
limited scope for investigating environmental aspects of the acquis in any detail. Our
approach was to focus on the capability of firms to measure emissions from their
production facilities. We sought to divide firms into those with emissions monitoring
equipment; those without it but aware that they should have it; and the final group that
saw no requirement to have monitoring equipment. There was a clear split between
Spanish firms and Eastern European ones (see Table 16 where size-corrected
distributions are presented). Just under half of the Spanish firms have monitoring
equipment with the rest stating they are not required to do so. No Spanish firms said
that they should but do not have equipment.
The presence of monitoring equipment is much sparser in Eastern Europe with no
discernible difference between Poland and Romania and no influence of ownership
type, corporate governance, pressure of competition or involvement in exporting. Less
than one-fifth of firms were able to monitor emissions and over seventy per cent stated
that there was no requirement for them to do so. A small but significant number of
firms in both countries stated that they should but did not have monitoring capability.
This comparison suggests that compliance is higher in Spain, it is better enforced and
that regulations are tougher.
4.6 Costs of compliance
It is extremely difficult to elicit reliable information from firms about the costs of
compliance with regulations. In many cases, managers will not be able to separate out
that component of an investment project or of training expenditure or of costs of
production that would have been saved had they not improved their level of
compliance. We have some information from firms for which managers reported that
some of their investment over the past three years had added to their standard of
compliance and who made an estimate of the proportion of investment spending that
would have been saved had they not sought to improve compliance. There are between
20 and 30 firms from each of Poland and Romania involved in each case. Estimates of
the proportion of investment that would have been saved are largeand should not be
taken too literally. For example, for each directive over half of the Romanian firms
reporting on the proportion of their investment that would have been saved had they
not improved compliance said that they would have saved more than ten per cent.
16
The results are quite consistent with the pattern of compliance outlined above. In
general, there was a tendency for Romanian firms to indicate a heavier burden than
Polish ones. Product certification appears to have been especially onerous for
Romanian firms with 24 out of the 27 firms involved reporting a saving of more than
10 per cent of their investment outlays had they not complied. It is worth recalling that
investment levels by Romanian firms were very low (especially amongst privatized
firms). It is notable that Polish and Romanian firms with emissions monitoring
equipment reported significantly higher levels of running costs than was true of the
Spanish firms.
4.7 Summary
Spain stands out from the Eastern European countries in its compliance with the
working time directive, its levels of ISO certification and emissions monitoring. Based
on the judgements of managers, Polish firms are indistinguishable from Spanish firms
and ahead of Romanian firms in occupational safety compliance. In terms of export
product certification, Poland stands out from Romania reflecting a deeper level of
integration awarenessthat includes firms that do not currently export to the EU. On
the harsher criterion of ISO certification, there is no discernible difference between the
East European countries. This confirms the emerging picture of a group of top-flight
privatized firms in Romania that are comparable to Polish ones but a longer tail of
firms for which the gap to EU standards is much more substantial. Only on equal pay,
is there no clear gap between the firms in Eastern Europe and in Spain.
From the survey results, it seems that transition indicators (ownership type and
corporate governance) and conditions of competition in the product market are not
decisive in explaining levels of compliance. Country differences dominate the picture of
compliance levels. A concern that at least some compliance levels may have been
driven by legacies from the pre-transition period with the consequence that new start-
ups would lag behind has not been confirmed by the survey results. Surprisingly the
levels of compliance by 1989 were low for state firms with adjustment there occurring
during transition. The only exception worthy of note is the lagging behind of Polish
start-up firms in meeting the working time directive.
5. Competing in a single market: preparedness and perceptions of
transition firms
In Sections 2 and 3, a picture was built up of the characteristics of firms in Poland and
Romania and they were compared with Spanish firms. On a number of dimensions of
performance, structure and compliance with EU directives, a clear gap remains
between Eastern European and Spanish firms. Yet there are dimensions on which
Polish firms look very similar to Spanish ones and quite different from Romanian ones.
In the compliance indicators, this is usually a country effect but in investment and
17
integration, private ownership both privatized and ab initio is often a correlate of
convergence with Spain. State-owned Polish firms, and all Romanian firms except new
private firms, tend to be laggards in adjustment. These results are quite reassuring since
progress in transition in the leading transition economy is producing measurable
convergence.
The attempt to probe more deeply the determinants of enterprise adjustment is
difficult. It has long been emphasized in the literature that selection issues in the
privatization process make the untangling of the ownership-corporate governance-
performance link very difficult. Using the survey data, we have not made much
headway in testing more directly for corporate governance effects. For example, we
were interested in whether there was any connection between firms in which there was
a new outside general manager (i.e., a general manager coming from outside the firm
but excluding cases where the general manager was present at the establishment of a
start-up) and levels or changes in performance. There appear to be few discernible
patterns. One exception was a weakly significant result for privatized Polish firms, for
which we found that a new outside general manager was more likely in firms for which
there was a single majority owner.
We also asked whether the present general manager had replaced a poorly performing
one. The affirmative answer was significantly more common in both Poland and
Romania than in Spain. This underlines the character of the transition process, even in
the early stage in Romania, as one in which there is an unusually high degree of
managerial turnover. Poor managerial performance can apparently be recognized and
replacement brought about. The conception that changes in the majority-owner of
firms might be the route through which effective corporate governance would be
executed does not receive much support in the data. The proportion of firms
experiencing a recent change in ownership was very similar in all three countries and
there seemed to be no correlation between a change of ownership and the presence of
a new outside general manager. However, it should be stressed that given the noisiness
of the data, there may simply be too few observations for hypotheses of this kind to be
confirmed or refuted with confidence.
In addition to the specific dimensions of convergence in terms of performance,
integration and compliance, we investigated the perceptions of firms about accession to
the EU and asked them to identify the most important impediments to the
implementation of their business plans. When asked for a subjective assessment of the
likely costs of accession, firms in Poland and Romania produced remarkably similar
responses. The first-ranked cost in both cases (identified by half the firms as the most
important) was the cost of compliance with EU regulations. Competition from the EU
ranked second (one-quarter of firms) and third was the fear of the loss of skilled labour
(about one in ten firms). Asked to assess the actual costs of accession, Spanish firms
pointed to competition from EU countries (one half of firms), the cost of compliance
18
to EU regulations (one-quarter) and competition from non-EU countries as the third
(one-fifth).
The benefits identified by firms in Poland and Romania were much more diffuse than
the assessment of costs and there was less concordance with the ex post views of the
Spanish firms. Managers of Polish and Romanian firms felt that accession would help
them to find a foreign partner, give them access to new technology, to cheap credit, to
a large market and assist because of the abolition of tariffs. Managers of Spanish firms
took the view that the main benefits were the abolition of customs formalities, the
future benefits of a common currency and access to a large market.
Features of the economic environment that businesses identify as obstacles to the
implementation of business plans can also provide useful information about the process
of adjustment. Managers were invited to rate the importance of each factor separately
before being asked to nominate the most important impediment. The factors they were
asked to consider were: delays and difficulties getting planning permission; problems
with legal permission or licenses to produce or operate; delays at customs for exports
or imports; difficulties in obtaining short-term or long-term bank credit, late payment
by domestic or foreign customers; problems with inconsistent taxation, changes in tax
regulations; too many different taxes; illegality and corruption; and environmental
regulation.
When asked to identify the most important obstacle, the Spanish firms in the sample
spread themselves rather widely across the possible answers. There was no particular
focus of discontent. By contrast, more than one-third of the Romanian and over 40%
of the Polish firms pointed to the late payment by domestic customers as the main
impediment to their business plans. The other focus for complaint in Eastern Europe
was the inconsistent, complex and unpredictable nature of taxation. (The interviewer
was specifically directed to emphasize that it was not the level of taxation that was at
issue.) It is striking that it was only the privatized and ab initio firms that identified
taxation as a serious problem for their firm. Problems with payments arrears may still
reflect the adjustment of firms in Eastern Europe to the characteristics of a market
economy rather than the existence of a much more serious problem in payment delays
than in Spain. It is notable, however, that the Polish firms were no less likely to identify
this as a major problem than the Romanian ones. The unpredictability of the tax
environment has been identified as a particularly serious problem for the countries that
are lagging in transition. The results here suggest that Polish private firms are very
concerned with this issue. Very few firms in any of the three countries identified
corruption as a major problem.
19
6. Conclusions
A major innovation of this study is to have used a market economy and member
country of the EU Spain as a benchmark against which to measure progress in
transition and readinessfor accession. As might have been expected, Spain is often
ahead on the measures of performance, integration and compliance that have been
examined but we have identified a number of dimensions on which Polish firms look
similar to Spanish ones. On the measures of integration and investment, it is Polish ab
initio and privatized firms that look most similar to the Spanish. Polish state-owned
firms often lag behind and look much more like Romanian firms than like either Polish
privatized or ab initio private firms. The fact that the investment share of Polish
privatized firms is higher than that of Spanish firms suggests that the ownership
structure and economic environment of these former state firms is able to support a
process of catch-up growth. We have also found evidence that there is a smaller gap in
Poland between some characteristics of firms that are already competing in the EU
market and firms that are only selling in the domestic market than is the case for
Romania. This provides another measurable indicator of the depth of integration and
hence of readinessfor accession since the presence of a large gap between these two
groups of firms would indicate that considerable adjustment was still required for the
firms operating only in the domestic market.
Taking a broad overview of compliance with EU directives, Poland tends to lag behind
Spain but lie significantly ahead of Romania. The Eastern European firms perceived a
heavy cost burden associated with raising compliance especially in connection with
emissions monitoring. A heavy burden was also identified by the Romanian firms in
relation to export product certification. There were fewer problems with complying
with equal pay, working time and for the Polish firms with occupational safety
directives. Levels of awareness of and compliance with directives did not vary with
ownership type amongst the Eastern European firms or with their involvement in
exporting. The hypothesis that compliance was legacy-based and hence more likely to
characterize state or formerly state-owned firms than start-ups was not confirmed. The
data available from this study suggest that compliance levels are fairly uniform within a
country. Progress in transition at the country level seems to be consistent with
improvements in compliance. At least from this study, there is little sign of a conflict
between the goals of transition toward a market economy and the improvement in
compliance with the major components of the acquis.
20
List of Tables and Figures
A guide to reading the tables
Table 1. Profitability 1997 (Earnings before interest, tax and depreciation as a
percentage of sales revenue.)
Table 2. Employment growth 199697, in per cent.
Table 3. Labour hoarding (percentage of firms reporting excess labour)
Table 4. Sales growth 199697, in per cent.
Table 5. Investment share 1997 (ratio of investment to sales, in per cent).
Figure 1. The distribution of exports as a percentage of sales across firms, by
country and ownership category.
Table 6. Quality of exported goods compared with those sold at home.
Table 7. Ownership concentration (distribution of firms according to whether they
have a single owner with a majority stake, at least one owner with a 20
50% stake, or dispersed ownership).
Table 8. Foreign ownership (percentage of firms with a majority foreign owner).
Table 9. Number of competitors in market for main product (percentage of firms
identifying none, between one and five, and more than five).
Table 10. How important is import competition? (percentage of firms identifying
imports as very important, average and not important as a source of
competition).
Table 11. Compliance with equal pay directive.
Table 12. Compliance with working time directive.
Table 13. Compliance with the occupational safety directive.
Table 14. Product certification requirements for exports to the EU.
Table 15. ISO9000 certification.
Table 16. Emissions monitoring equipment.
Appendix
Table A1. Basic data on survey firms.
21
Tables and Figures
How to read these tables:
Tables with statistical tests are based on regressions with the firm characteristic of
interest as the dependent variable, and country/ownership dummy variables and log
employment in 1997 as the independent variables. Country/ownership categories (7):
Spain, Poland state-owned, Poland privatized, Poland ab initio private, Romania state-
owned, Romania privatized, Romania ab initio private. Log employment was included
as a simple control for the size of the firm.
The regression technique chosen follows from the type and nature of the dependent
variable.
Two sets of statistical tests are usually presented:
(1) Country by country tests. These are pairwise tests of whether the two countries in
question are different, in the sense of a joint test of all dummies for one country being
significantly different from all dummies for the second country.
9
These are shown in
the country rows and columns.
(2) Country/ownership category tests. These are pairwise tests of whether the firms of
one country-ownership combination (e.g., Polish state-owned) have a higher or lower
value than the firms of the other country-ownership combination (e.g., Romanian
privatized, or all Spanish).
10
These are shown in the country-ownership rows and
columns. A +sign means the firms of that row category have on averagea larger value
of the characteristic in question than the firms in the column category; a sign means a
lower value for the row category; zero means no statistical significance.
All tests are at the 5% significance level.

9
For example, the test of whether Poland is different from Spain is a joint test of whether the Polish
state-owned, privatized and ab initio dummies are significantly different from the Spain dummy; the
test of whether Romania is different from Poland is a joint test of whether the Polish and Romanian
state-owned dummies are different, whether the two privatized dummies are different, and whether
the two ab initio dummies are different.
Table 1. Profitability 1997 (Earnings before interest, tax and depreciation as a percentage of sales revenue)
Means
(Medians)
Means
controlling
for size
Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 11.4
(10.9)
11.3
Poland 11.6
(10.2)
Diff
Romania
(1996)
14.1
(12.9)
Diff Diff
Poland:
SOE 10.1
(9.2)
10.4 0
Privatized 9.6
(8.5)
9.7 0 0
Ab initio 15.6
(13.6)
15.4 + + +
Romania:
(1996)
SOE 10.9
(11.5)
11.8 0 0 0
Privatized 14.4
(12.7)
14.8 + + + 0 +
Ab initio 15.5
(13.7)
14.8 + + + 0 0 0
Notes: Corrected for inflation effects (see text). Statistical tests use OLS. Means controlling for size are predicted values for a firm with 250 employees.
24
Table 2. Employment growth 199697, in per cent
Means
(Medians)
Means
controlling
for size
Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 4.8
(1.7)
4.7
Poland 0.5
(0.7)
Diff
Romania
(199596)
0.8
(2.2)
Diff No diff
Poland:
SOE 5.2
(3.5)
4.7
Privatized 6.4
(5.5)
6.2 0
Ab initio 13.6
(9.3)
13.4 + + +
Romania:
(199596)
SOE 7.1
(5.2)
6.3 0 0
Privatized 7.7
(4.8)
7.3 0 0 0
Ab initio 14.7
(7.3)
14.0 + + + 0 + +
Note: Log growth rates. Statistical tests use OLS. Means controlling for size are predicted values for a firm with 250 employees.
25
Table 3. Labour hoarding (L) (percentage of firms reporting excess labour)
Percentage
of firms
reporting
excess L
Same,
controlling
for size
Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 16.7 16.8
Poland 44.8 Diff
Romania 22.3 Diff Diff
Poland:
SOE 65.9 62.9 +
Privatized 51.4 50.5 + 0
Ab initio 24.0 24.6 0
Romania:
SOE 39.5 34.8 + 0 0
Privatized 24.4 22.1 0 0 0
Ab initio 11.1 12.8 0 0 0
Note: Statistical tests use logit with dependent variable = 1 if firm reported excess labour, 0 otherwise. Percentages controlling for size are predicted probabilities for a firm
with 250 employees.
26
Table 4. Sales growth 199697, in per cent
Means
(Medians)
Means
controlling
for size
Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 9.5
(7.7)
9.3
Poland 3.6
(2.1)
Diff
Romania
(199596)
8.4
(6.7)
Diff No diff
Poland:
SOE 2.8
(2.7)
2.0
Privatized 1.2
(1.7)
1.6 0
Ab initio 11.2
(9.7)
10.9 0 + +
Romania:
(199596)
SOE 0.4
(3.8)
1.2 0 0
Privatized 4.1
(5.2)
5.0 0 0 0 0 0
Ab initio 18.2
(16.6)
16.9 + + + 0 + +
Note: Log growth rates; sales deflated using official PPI (Poland, Spain) or implicit industrial deflator (Romania, see text for details). Statistical tests use OLS. Means
controlling for size are predicted values for a firm with 250 employees.
27
Table 5. I nvestment share 1997 (ratio of investment to sales, in per cent)
Means
(Medians)
(Medians
controlling for size)
Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 5.0
(2.0) (2.0)
Poland 9.5
(4.3) Diff
Romania
(1996)
6.0
(4.0) Diff Diff
Poland:
SOE 5.1
(2.0) (2.0) 0
Privatized 11.2
(4.8) (6.0) + +
Ab initio 9.4
(5.0) (6.0) + + 0
Romania:
(1996)
SOE 4.6
(4.0) (4.0) 0 0 0
Privatized 5.9
(2.4) (4.0) 0 0 0
Ab initio 6.9
(5.0) (6.0) + + 0 0 0 +
Note: Statistical tests use ordered logit with dependent variable as investment/sales ratio = 0%, 1%, 2%, . etc.
Column controlling for size derives from predicted probabilities for each investment/sales category for a firm of 250 employees;
value is predicted median category of investment/ratio.
28
Figure 1. Percentage of sales exported
Notes: P = Poland, R = Romania, SOE = State-owned enterprise, Pr = Privatized enterprise, AI = Ab initio private firm.
Export share categories are: 0%, 110%, 1120%, , 9099%, 100%.
29
Table 6. Quality of exported goods compared with those sold at home
Number
of firms
Percentage of row category
Export
quality
higher
Export
quality
same
Export
quality
lower
Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 197 0 99 1
Poland 125 2 96 2 Diff
Romania 95 16 84 0 Diff Diff
Poland:
SOE 28 7 93 0
Privatized 62 2 97 2 0 0
Ab initio 35 0 97 3 0 + 0
Romania:
SOE 22 27 73 0 0
Privatized 51 14 86 0 0 0
Ab initio 22 9 91 0 0 0 0 0
Note: Statistical tests use ordered logit with dependent variable = 0 if export quality is higher than the same product sold domestically, = 1 if export quality is the same, = 2
if export quality is lower. A entry indicates, e.g., that the shortfall in quality of goods sold domestically compared to export quality is greater in the row category than it
is in the column category. Regression includes log employment as a control for size effects. NB: Table refers only to firms selling same product on export and domestic
markets.
30
Table 7. Ownership concentration (distribution of firms according to whether they have a single owner with a majority stake, at least one
owner with a 2050% stake, or dispersed ownership)
Ownership concentration
Percentage of row category
Dispersed At least one
owner with a
2050% stake
Single owner
with majority
stake
Spain Poland Poland
Privatized
Poland
Ab initio
Romania
Privatized
Spain 16% 25% 59%
Poland 36% 31% 33% Diff
Romania 37% 28% 36% Diff No Diff
Poland:
Privatized 53% 22% 25%
Ab initio 11% 43% 44% 0 +
Romania:
Privatized 61% 18% 21% 0
Ab initio 12% 37% 51% 0 + 0 +
Notes: Statistical tests use ordered logit with dependent variable = 0 if dispersed ownership, 1 if at least one owner with 2050% stake,
2 if firm has single private owner with majority stake. Regression includes log employment as a control for size effects. NB: Private firms only.
31
Table 8. Foreign ownership (percentage of firms with a majority foreign owner)
Domestic
Number
(Per cent)
Foreign
Number
(Per cent)
Total
Number
(Per cent)
Spain 125
(58%)
90
(42%)
215
(100%)
Poland
of which:
194
(87%)
29
(13%)
223
(100%)
SOE 41
(100%)
0
(0%)
41
(100%)
Privatized 92
(86%)
15
(14%)
107
(100%)
Ab initio 61
(81%)
14
(18.7%)
75
(100%)
Romania
of which:
195
(94%)
12
(5.8%)
207
(100%)
SOE 44
(100%)
0
(0%)
44
(100%)
Privatized 76
(93%)
6
(7%)
82
(100%)
Ab initio 75
(93%)
6
(7%)
81
(100%)
32
Table 9. Number of competitors in market for main product (percentage of firms identifying none, between one and five, and more than
five)
Number of competitors
Percentage of row category
None 15 > 5 Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 16 47 37
Poland 23 30 47 No Diff
Romania 14 27 60 Diff Diff
Poland:
SOE 17 24 59 +
Privatized 23 33 44 0 0
Ab initio 27 28 45 0 0
Romania:
SOE 34 30 36 0 0 0
Privatized 11 35 54 + 0 + + +
Ab initio 5 16 79 + 0 + + + +
Note: Statistical tests use ordered logit with dependent variable = 0 if firm dominates market, 1 if firm reports 15 competitors, 2 if firm reports > 5 competitors.
Regression includes log employment as a control for size effects.
33
Table 10. How important is import competition? (percentage of firms identifying imports as very important, average and not important as a
source of competition)
Import competition
Percentage of row category
Not
impt.
Avg. Very
impt.
Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 36 43 21
Poland 28 35 37 Diff
Romania 29 35 36 Diff No Diff
Poland:
SOE 24 39 37 0
Privatized 22 35 43 + 0
Ab initio 37 35 28 0 0
Romania:
SOE 34 34 32 0 0 0 0
Privatized 24 44 32 + 0 0 0 0
Ab initio 31 27 42 + 0 0 0 0 0
Note: Statistical tests used ordered logit with dependent variable = 0 if not important, 1 if intermediate importance, 2 if very important. Regression includes log
employment as a control for size effects.
34
Table 11. Compliance with equal pay directive
Extent of compliance
1 = very low
7 = high compliance
Percentage of row
category
5 or
less
6 7 Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 3 3 94
Poland 3 5 92 No Diff
Romania 1 1 98 No Diff No Diff
Poland:
SOE 5 10 85 0
Privatized 5 3 92 0 0
Ab initio 1 4 95 0 0 0
Romania:
SOE 0 0 100 n.a. n.a. n.a. n.a.
Privatized 2 2 96 0 + 0 0 n.a.
Ab initio 1 0 99 0 + 0 0 n.a. 0
Note: Statistical tests use an ordered logit estimation with dependent variable = extent of compliance. Statistical tests not available for Romanian SOEs because of 100%
compliance. Regression includes log employment as a control for size effects.
35
Table 12. Compliance with working time directive
Extent of compliance
1 = very low
7 = high compliance
Percentage of row
category
5 or
less
6 7 Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain < 1 4 96
Poland 3 13 84 Diff
Romania 4 21 75 Diff Diff
Poland:
SOE 2 10 88
Privatized 2 10 88 0
Ab initio 4 20 76 0
Romania:
SOE 7 20 73 0 0
Privatized 2 22 76 0 0 0
Ab initio 4 21 75 0 0 0 0
Note: Statistical tests are from an ordered logit estimation with dependent variable = extent of compliance. Regression includes log employment as a control for size effects.
Difference between statistical test results in Poland SOE column and Poland privatized column derive from smaller standard errors for latter (resulting from larger number
of Polish privatized firms in the sample).
36
Table 13. Compliance with occupational safety directive
Extent of compliance
1 = very low
7 = high compliance
Percentage of row
category
5 or
less
6 7 Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 4 12 84
Poland 4 13 83 No Diff
Romania 11 32 57 Diff Diff
Poland:
SOE 5 12 83 0
Privatized 4 16 80 0 0
Ab initio 5 8 87 0 0 0
Romania:
SOE 18 41 41
Privatized 10 33 57 0
Ab initio 10 26 64 0 + 0
Note: Statistical tests are from an ordered logit estimation with dependent variable = extent of compliance. Regression includes log employment as a control for size effects.
37
Table 14. Product certification requirements for exports to EU
Familiarity with product certification
requirements for exports to EU
Percentage of firms in row category
No
knowledge
Some
knowledge
Very
familiar
Poland Poland
0%
Poland
124%
Poland
25%+
Romania
0%
Romania
124%
Poland 18 28 54
Romania 39 20 41 Diff
Poland:
No EU exports 22 28 50
124% of sales 13 33 54 0
25% + of sales 19 17 64 0 0
Romania:
No EU exports 64 15 21
124% of sales 12 27 62 0 0 0 +
25% + of sales 9 26 65 0 0 0 + 0
Note: Statistical tests are from an ordered logit estimation with dependent variable = familiarity with requirements. Regression includes log employment as a control for
size effects.
38
Table 15. I SO9000 Certification
Percent
of firms with
ISO900
Same,
controlling
for size
Spain Poland Poland
SOE
Poland
Privatized
Poland
Ab initio
Romania
SOE
Romania
Privatized
Spain 61 64
Poland 21 Diff
Romania 20 Diff No Diff
Poland:
SOE 20 13
Privatized 26 22 0
Ab initio 15 14 0 0
Romania:
SOE 30 19 0 0 0
Privatized 22 16 0 0 0 0
Ab initio 14 18 0 0 0 0 0
Note: Statistical tests are from logit estimation with dependent variable = has some ISO certification. Column controlling for size is predicted percentages for a firm of 250
employees. NB: Firms with ISO 9001, 9002 or 9003.
Table 16. Emissions monitoring equipment
Does the firm have emissions monitoring equipment?
Percentage of row category
(Percentage, size-corrected)
Yes Should have but
doesnt
No requirement to
have equipment
Spain 44
(45)
0
(0)
56
(55)
Poland 14 5 81
Romania 19 10 71
Poland:
SOE 20
(13)
2
(2)
78
(85)
Privatized 16
(13)
7
(7)
77
(80)
Ab initio 8
(8)
3
(3)
89
(90)
Romania:
SOE 30
(20)
16
(14)
55
(65)
Privatized 21
(15)
9
(8)
71
(77)
Ab initio 12
(16)
7
(9)
80
(75)
Note: Statistical tests (not reported above) used multinomial logit with dependent variable = yes,
should, no requirement. Size-corrected percentages are predicted probabilities from the regression.
40
Appendix
Table A1: Basic data on surveyed firms
Spain Poland Romania
Number of firms 215 223 207
Size distribution (% of total)
50 to 150 employees 42 32 36
151 to 500 emloyees 38 36 33
501 to 5000 employees 20 32 31
Sectoral distribution (% of total)
Mechanical engineering and transport 12 15 14
Wood 4 4 13
Food processing 7 14 15
Clothing and textiles 11 12 21
Chemicals 19 11 5
Electrical machinery 12 14 6
Miscellaneous/combination of above 35 30 26
Ownership (% of total)
State-owned n.a. 18 21
Privatized (previously state-owned) n.a. 48 40
Ab initio (new private firm) n.a. 34 39
Mean (median) 1997 employment by ownership
All 463
(181)
464
(283)
490
(230)
State-owned n.a. 637
(475)
799
(597)
Privatized (previously state-owned) n.a. 479
(305)
620
(393)
Ab initio (new private firm) n.a. 347
(190)
191
(110)